There are serious pitfalls for investors when trying to assess company earnings.
Since Tesco (LSE: TSCO) recently released its results for the year ended 25 February 2012, investors -- including many Fools -- have been trying to get to grips with its earnings.
Maybe you're looking at Tesco's earnings for the first time, having simply followed Warren Buffett's lead, and bought shares in the company after its profit warning in January.
(By the way, you can discover what price Buffett paid for his Tesco shares in this free report!)
Maybe you're a longstanding shareholder and are puzzled why Tesco's earnings per share (eps) seems to routinely come in a couple of pence higher than analysts had been forecasting.
What's going on? What are Tesco's 'real' earnings?
A multitude of numbers
If you go to the Income Statement in Tesco's latest accounts, you'll find no fewer than five different eps numbers, as follows:
| Earnings per share from continuing and discontinued operations | |
|---|
| Basic | 34.98p |
| Diluted | 34.88p |
| | |
| Earnings per share from continuing operations | |
| Basic | 36.75p |
| Diluted | 36.64p |
| | |
| Underlying diluted earnings per share from continuing operations | 37.41p |
Basic eps amounts are calculated using the number of shares in issue, while diluted eps also takes account of share options granted to directors and employees. In Tesco's case, as with most large companies, the difference is minimal.
The first two numbers in the table give us basic and diluted eps for the whole business. However, Tesco put its loss-making Japanese business up for sale during the year, so the second two numbers gives us basic and diluted eps as if the Japanese operation didn't exist.
Finally, we get an 'underlying' diluted eps number. This excludes not only the Japanese operation, but also certain non-trading and one-off items. The directors believe excluding these items gives "additional useful information for shareholders on underlying trends and performance".
Comparing apples with apples
It's not mandatory for companies to provide underlying eps and there are no absolute rules for calculating it. Nevertheless, most companies do give such a number. Furthermore, this underlying -- also called 'adjusted' or 'normalised' -- eps is the one many investors rely on when calculating the historic version of the popular price-to-earnings (P/E) valuation ratio.
Now, there's a crucial thing a lot of investors appear to be unaware of. Any analyst worth his or her salt doesn't just blindly accept the underlying eps provided by a company. Analysts do their own calculations -- and for a good reason. Tesco actually gives the reason in Note 1 of its accounts: the underlying figure "may not be directly comparable with other companies' adjusted profit measures".
By doing his or her own calculations using a consistent set of rules, an analyst can be sure of having underlying eps numbers that are directly comparable. Thus, earnings-based valuations, such as P/E, are also on an apples-with-apples basis, and a comparison of the P/Es of, say, Tesco, Sainsbury (LSE: SBRY) and Morrison (LSE: MRW) becomes meaningful.
Putting Tesco to the test
The problem for many private investors is they don't have access to analysts' historic figures, or the time or knowledge to do their own normalised eps calculations.
Most of the free data providers used by private investors simply plug the underlying eps number given by a company into their databases, which means investors are often unwittingly comparing apples with oranges when using trailing P/Es to judge the relative merits of companies.
One exception is Morningstar, whose analysts do actually go to the effort of doing their own calculations of normalised eps.
Let's look at the company-provided and Morningstar-provided eps numbers for Tesco and the UK's other two listed supermarkets on a trailing 12-month basis, and see how the P/Es come out:
| Company | Current share price (p) | Company normalised eps (p) | P/E | Morningstar normalised eps (p) | P/E |
|---|
| Tesco | 317.95 | 37.41 | 8.5x | 33.26 | 9.6x |
| Sainsbury | 307.20 | 26.90 | 11.4x | 28.13 | 10.9x |
| Morrison | 283.20 | 24.93 | 11.4x | 26.11 | 10.8x |
On historic P/Es derived from company-provided eps figures -- which, as we know, may not be directly comparable -- Tesco looks wildly undervalued against its peers. However, on the Morningstar figures, which we can assume have been calculated using a consistent methodology, the difference is considerably less.
As seems to be the case with analysts' generally, Morningstar's normalised eps numbers for Tesco are persistently below those given by the company itself. This is presumably the result of analysts using methodologies for calculating underlying eps that enable a like-for-like comparison of Tesco with its peers.
The growth forecast pitfall
Another serious pitfall for investors, flowing from the same issues, is misinterpreting forecast eps growth.
Let's have a look at the current analysts' consensus forecasts for Tesco, and see how eps growth comes out when taken, on one hand, from the base of the company-provided 37.41p historic eps and, on the other, from the base of the Morningstar 33.26p historic eps.
| Consensus forecast eps* | Eps growth from 37.41p | Eps growth from 33.26p |
|---|
| 34.7p | -7% | +4% |
* Aggregated from several data providers
Given what the directors have said about the prospects for the year ahead, the analysts' consensus forecast of 34.7p doesn't make much sense against Tesco's historic 37.41p, because it gives a substantial 7% fall in eps. It makes much more sense against Morningstar's historic 33.26p, which gives a 4% increase. (I believe Morningstar's 33.26p is probably a little below consensus on this occasion and the consensus for eps growth is actually nearer 1%-2%.)
Foolish bottom line
Underlying eps figures provided by companies are often calculated in different ways. As such investors may be unwittingly comparing apples with oranges when using historic P/Es to judge the relative merits of companies.
In addition, because analysts do their own calculations of underlying eps, investors who compare analysts' consensus forecast eps with the historic eps provided by companies may arrive at a seriously flawed conclusion about expected eps growth.
Morningstar's numbers give investors some protection against these pitfalls. In my experience, they're often reasonably close to the analysts' consensus. However, they can sometimes be a long way off, so you do need to exercise caution and common sense.
For help trying to grow and preserve your wealth, start your 30-day free trial to Motley Fool Share Advisor today. You'll receive immediate access to what we believe are the best shares you can buy right now, complete with our in-depth analysis and ongoing coverage.
Further investment opportunities:
